When an error made a record member invalid, the record would stay as "isBeingDefined" and
not "completeDefinition". Even easily recoverable errors ended up propagating records in
such "beingDefined" state, for example:

struct A {
  ~A() const; // expected-error {{'const' qualifier is not allowed on a destructor}}
};
struct B : A {}; // A & B would stay as "not complete definition" and "being defined".

This weird state was impending lookups in the records and hitting assertion in the ASTWriter.

Part of rdar://11007039

llvm-svn: 152432

llvm-svn: 152431

llvm-svn: 152430

llvm-svn: 152429

LLVM_READONLY.

llvm-svn: 152428

Handle the case where we get called to determine the ObjC runtime version BEFORE the loader code has
winnowed all the unloaded libraries from the process module list.
<rdar://problem/11015223>

llvm-svn: 152427

 - The theory here is that we have these functions sprinkled in all over the
   place. This should allow the optimizer to at least realize it can still do
   load CSE across these calls.
 - I blindly marked all instances as such, even though the optimizer can infer
   this attribute in some instances (some of the inline ones) as that was easier
   and also, when given the choice between thinking and not thinking, I prefer
   the latter.

You might think this is mere frivolity, but actually this is good for a .7 -
1.1% speedup on 403.gcc/combine.c, JSC/Interpreter.cpp,
OGF/NSBezierPath-OAExtensions.m.

llvm-svn: 152426

to count the number of phis changed, not the number visited.

llvm-svn: 152425

llvm-svn: 152424

Fixing a problem with the NSSet summary provider; plus, providing a further optimization to the whole Cocoa formatters infrastructure

llvm-svn: 152423

Previous code had a double free in MemoryBuffer. The tests now pass.

llvm-svn: 152422

traversal, consider nodes for which the only successors are backedges
which the traversal is ignoring to be exit nodes. This fixes a problem
where the bottom-up traversal was failing to visit split blocks along
split loop backedges. This fixes rdar://10989035.

llvm-svn: 152421

to forward class, and assigning to an 'id' type var, message
sends default to 'id'. // rdar"//10988847

llvm-svn: 152420

 - getSourceRange().getBegin() is about as awesome a pattern as .copy().size().

I already killed the hot paths so this doesn't seem to impact performance on my
tests-of-the-day, but it is a much more sensible (and shorter) pattern.

llvm-svn: 152419

llvm-svn: 152418

llvm-svn: 152415

prefix.  Added a FIXME to remind us this still does not work when it is not the
first prefix.

llvm-svn: 152414

names.

llvm-svn: 152413

 - This cuts the # of getSourceRange calls by 60% on
   OGF/NSBezierPath-OAExtensions.m.

llvm-svn: 152412

 - getSourceRange() can be very expensive, we should try to avoid it if at all possible.

In conjunction with the previous commit I measured a ~2% speedup on 403.gcc/combine.c and a 3% speedup on OmniGroupFrameworks/NSBezierPath-OAExtensions.m.

llvm-svn: 152411

 - There are probably a lot more of these worth adding, but these are a start at hitting some of the exprs for which getSourceRange().getBegin() is a poor substitute for getLocStart().

llvm-svn: 152410

llvm-svn: 152409

more obvious.

llvm-svn: 152408

~0U might be i32 on 32-bit hosts, then (uint64_t)~0U might not be expected as (i64)0xFFFFFFFF_FFFFFFFF, but as (i64)0x00000000_FFFFFFFF.

llvm-svn: 152407

llvm-svn: 152406

negative switch cases if the branch condition is known to be positive.
Inspired by a recent improvement to GCC's VRP.

llvm-svn: 152405

llvm-svn: 152404

This renames the -Wformat-non-standard flag to -Wformat-non-iso,
rewords the current warnings a bit (pointing out that a format string
is not supported by ISO C rather than being "non standard"),
and adds a warning about positional arguments.

llvm-svn: 152403

llvm-svn: 152401

Patch by Tom Stellard!

llvm-svn: 152400

it to analyze extractvalue(llvm.[us](add|sub).with.overflow.*) intrinsics!

llvm-svn: 152398

This contains a semi-functional skeleton for the implementation of the
LLVM bindings for Python.

The API for the Object.h interface is roughly designed but not
implemented. MemoryBufferRef is implemented and actually appears to
work!

The ObjectFile unit test fails with a segmentation fault because the
LLVM library isn't being properly initialized. The build system doesn't
know about this code yet, so no alerts should fire.

llvm-svn: 152397

llvm-svn: 152396

llvm-svn: 152395

llvm-svn: 152394

llvm-svn: 152393

and lots of tidying up.

llvm-svn: 152392

llvm-svn: 152391

llvm-svn: 152390

llvm-svn: 152389